Summary While upbound in daylight and in fair weather at around 1422, the Cedar, under conduct of a pilot, lost rudder control off Deschaillons-sur-Saint-Laurent and grounded on the north side of the channel near buoyD-36. While the fore part of the Cedar was hard aground, the after half of the vessel remained in the channel. After sounding the tanks, it was found that the forepeak and the No.1 double bottom ballast tank on the starboard side were holed. It was decided to increase the aft trimming moment by ballasting the afterpeak and other ballast tanks located aft. Later that evening, with the high tide, the vessel was freed under its own power. It was then conducted to the Port of Qubec, without further steering problems, for inspection and repairs. No injuries or pollution resulted from this occurrence. Ce rapport est galement disponible en franais. Other Factual Information Particulars of Vessel Description of the Vessel The Cedar is a conventional bulk carrier with five holds and four cranes (Photo1). The vessel was constructed under Lloyd's Register (LR) specifications and standards at the Wuhu Shipyard, China, to maximum St.Lawrence Seaway dimensions. At the time of the grounding the vessel remained classed with LR. The vessel is equipped with a course and rudder angle recorder that gives a printout of courses steered and rudder angle achieved on a common time scale. Other data available from different sources on board include engine telegraph orders and an alarm log - both automatically monitored and each recorded on its own paper printout. The wheelhouse steering position is equipped with a follow-up2 (FU) wheel and a non-follow-up3 (NFU) joystick and push button arrangement. There is also another NFU joystick on the main control panel of the wheelhouse. These control mechanisms are part of integrated C.Plath package that includes, among others, the telemotor transmitter, autopilot system, gyro and magnetic compasses. History of the Voyage On 20 September 2001, the Cedar left the Wuhu shipyard on her maiden voyage, stopping first at Fangcheng, China. The vessel was loaded with bauxite, brown alumina grude and brown alumina dregs to a draught of 9.27m forward and 9.64m aft. The vessel departed Fangcheng on 30September, bound for Qubec, Quebec and Thorold, Ontario via the Panama canal. After a short stop at Hong Kong for bunkers, the vessel made the Pacific crossing and then transited the Panama Canal between October31 and November3. On November13, the vessel was off Les Escoumins pilot station in the St.Lawrence River where a river pilot boarded. As the vessel proceeded up river under the conduct of a pilot, both steering gear pumps were on line. Steering was carried out in manual mode from the steering station using the FU wheel. No steering anomaly had been noted since departing China. The vessel stopped at Qubec on the evening of November13 to unload a portion of the brown alumina grude and obtain an acceptable Seaway draught of 7.98m fore and aft. The Cedar departed Qubec at 0955 on November16. Passage upriver was without incident, and the vessel was making about 9 knots over the bottom, against the river flow and ebbing tide of approximately 2-3knots. On the bridge at this time were the officer of the watch (OOW), the helmsman, a pilot and an apprentice pilot. Both steering pumps were on line and the helmsman was using the FU wheel. At about 1418, when the vessel was near Deschaillons-sur-Saint-Laurent, a course alteration to port was given and the helmsman brought the vessel onto 257(G)(position1, Figure1)4. Then, to complete the turn, the pilot ordered 248, and then 246, in quick succession. The helmsman put on 15of port helm and the vessel's head shifted quickly to port, overshooting the requested 246and stopping at about 238(position2, Figure1). By now, the helmsman had applied starboard counter-rudder (approximately 19) and the ship's head began to come back to starboard. He then put the helm first 10then 20to port in order to stop the swing at the required 246. The vessel continued its starboard swing and, at approximately 1420, the helmsman realized that the rudder was not responding to port helm commands. The pilot ordered hard to port, but there was still no response from the rudder. In very quick succession, both NFU joysticks were tried but neither gave any effect. The vessel was swinging to starboard and the current, setting at about 070(T), had caught the port bow. At about this time, the master entered the bridge and was informed of the loss of steering. Full astern was ordered on the main engine and the crash stop sequence was initiated. Before astern power could be developed, the vessel exited the north side of the channel on a heading of approximately 285. The Cedar was still making close to 9knots and, as the bow came into contact with the channel bottom, the vessel's head shifted to port about 7before coming quickly to a stop at 1422. The vessel was now aground at position 4633'48N; 07207'33W on a heading of 278(position3, Figure1), the helm hard to port, but the rudder still to starboard about 4. Figure1. Ship's movement (positions approximate) Photo2. Emergency Remote Control Transfer Switch Shortly after the grounding, the ship's electrician went to the steering gear flat and isolated the steering gear control from the wheelhouse using the emergency remote control transfer switch (Photo2). He then tried the rudder from the emergency steering position using the NFU push buttons on either side of the switch. The rudder responded normally and after a few tests, he turned the emergency remote control transfer switch to the wheelhouse position. Subsequent tests from the wheelhouse confirmed the rudder was now following FU and NFU commands. Tanks were sounded and it was found that the forepeak and No.1 starboard double bottom ballast tank were holed. The pilot and master agreed that a refloating would be attempted at the next high tide without tug assistance. Calculations were done on the vessel's loading computer for ballasting the afterpeak and other aft ballast tanks. The ballasting scenario resulted in a maximum bending moment of 88% of maximum at frame111. At 1930, two Transport Canada Marine Safety (TCMS) inspectors arrived on board. Various steering gear tests were undertaken and everything seemed to be working normally. At about 2100, near high tide, the vessel began to pivot to starboard - astern power was applied but the vessel remained aground. At about 2130, after nearly 30minutes of continuous astern power application, and with the ebb tide about to begin, the vessel floated free. It was then backed into the channel and conducted down river to the Port of Qubec for further inspections and temporary repairs. Steering Gear Arrangements The steering gear arrangements on the Cedar include a Rapson slide electro hydraulic steering gear, manufactured by Kawasaki (TypeRM21-040). The rudder actuator is composed of two rams supplied by two variable-delivery pumps (main steering gear pumps), each with its own corresponding electric motor. The main pumps are operated through a floating lever arrangement that is actuated by one (orboth) hydraulic telemotor receivers. Each telemotor receiver consists of a servo power unit and cylinder. The servo cylinders are mounted axially opposed to each other and connected to the floating lever at a common point (Figure2). Figure2. Steering gear arrangements The servo power units and their associated electrical arrangement were supplied by the shipyard. Each unit comprises an electric motor, hydraulic pump, control group valve chest as well as connecting hydraulic pipe. The control group valve chest of each servo power unit includes a solenoid-operated directional control valve (DCV) and a hydraulically-operated servo cylinder bypass valve. The DCV solenoid receives a low voltage (24Vdc) control signal from the telemotor transmitter. The servo cylinder bypass valve is held in the bypass condition by a spring when system pressure is zero (servo pump off). When the servo pump is activated, the pressure builds up and overcomes the spring's force, closing the bypass and hydraulically locking the associated servo cylinder. After the grounding it was found that, although both port and starboard bypass valves were in place on the control group valve chest of each servo power unit, they had been incorrectly installed (spring for spool valve on same side as servo pump pressure).5 The port and starboard servo power units were connected electrically with their corresponding main steering gear pump (port or starboard) such that when a particular steering gear pump was selected (from the wheelhouse), the associated servo power unit was automatically activated. When both main steering gear pumps were selected for simultaneous operation, the two servo power units worked in tandem. Each servo power unit has its own electrical starter panel, supplied by the shipyard during construction. Each starter panel comprises a motor contactor, various fuses and a transformer. After the grounding, the main power switch within the starter panel of the port side servo power unit, although still functional, was found in a state of partial disassembly. One of two bolts holding the switch assembly together had released. During the course of the investigation, a loose wire was found within the emergency remote control transfer switch in the steering gear flat (Photo2). This wire was the low voltage control signal to the starboard servo power unit DCV solenoid for port rudder activation. Other wires within this switch were also found loose. The wires exhibited signs of being crimped with a tool other than a proper crimping tool. Control and Operation of the Steering Gear The servo cylinders receive oil via their respective directional control valves. When control input is applied from any source using FU or NFU methods, two distinct 24Vdc control signals are sent, one to each DCV solenoid on the port and starboard servo power units. Consequently, both DCVs operate at all times, regardless if one or both servo power units is actually in use. Apart from the FU and NFU steering positions in the wheelhouse, there are two steering positions located within the steering gear flat. The first is a NFU arrangement where control is transferred to the steering gear flat by way of an emergency remote control transfer switch. Steering is then effected by pressing the NFU buttons just above the switch. This switch has a protective latch to prevent accidental selection of the emergency position. The second method, in the event of failure of both servo power units, manually activates the main steering pump(s) via the floating lever using the trick wheel, just forward of the steering gear.6 To do so, however, the trick wheel must be engaged in the floating lever by inserting a bolt in the specified position. As the remote steering control system is electric, an audible and visual alarm on the bridge is provided in case of electrical power supply failure for this system.7 During the loss of steering event, no alarms were heard by any of the navigation team, nor was there any such alarm logged by the alarm log in the engine control room. Servo Power Unit Hydraulic Piping When installed at the shipyard, the connecting piping of the servo power units was copper. The piping connections were installed using direct braising methods and simple butt joints (Photo3). Joints contained much excess braising material on the exterior, and the one examined in detail during the course of the investigation showed an overflow of braising material into the internal bore (Photo4). It is estimated that the obstruction would restrict flow to about 70% of the design rate.8 Bends in the piping were unsymmetrical and flattened at points to the extent that, during repairs subsequent to the grounding, TCMS required all servo power unit piping be replaced with steel piping, properly bent, fitted, and sized. Photo3. Piping connections using direct braising and butt joints Photo4. Overflow of raising material into the internal bore of piping Sister Ship Constructed at the same shipyard, the Pine was delivered to the Greek operator, Diana Shipping Agencies, five months before the Cedar. The Pine was visited subsequent to the grounding of the Cedar and found to have identical steering gear arrangements to the Cedar except for the following: the main power switch for each servo power unit was intact; the hydraulic bypass valve for each servo cylinder was correctly installed; the bending and brazing work on the copper hydraulic piping of the servo-power units appeared somewhat better than that seen on the Cedar. Trick Wheel Operating Instructions Emergency steering instructions posted in the steering gear flat on the Cedar (as well as the Pine) were engraved on two bronze plaques installed one above the other. Each plaque was intended to explain the use of the trick wheel and read as follows: In case of handling the trick wheel, insert the pin A into the position for emergency. In other cases keep the pin A releasing from the nut. Don't draw out the pins B and C. Disconnect the remote steering gear control system from the power circuit. Shift the pin of trick wheel to the position of for emergency. Operate the power unit by means of E/M start switch. Trick wheel operation is necessary in the event of servo power failure. Incorrect installation of the servo power unit bypass valves on the Cedar meant that even when the servo pumps were off (or not functional), the associated servo cylinders were still hydraulically locked instead of in the bypass condition. In order for the trick wheel to operate the floating lever under these conditions, it would now require the extra step of decoupling both servo cylinders from the floating lever. Damage to the Vessel The forepeak and No.1 starboard double bottom ballast tanks were punctured between frames216 and235 (approximately 12m in length) running from the keel plate to the starboard bilge. The shell plating in way of damage was set in approximately 600mm. The duct keel plate in way of frames175 to185 was also set in. Other bottom plates were set in as far astern as frame151 and affected tanks 1P, 2PS, 4PS, 5PS, and 7S. Photo6. No.1 starboard bottom ballast damage